This proposal describes a detailed plan for a Cancer Genome Characterization Center (CGCC) to act as part of The Cancer Genome Atlas (TCGA) project. This CGCC will contribute two platforms to accurately analyze over 1,000 cancer genomes per year using two of the most efficient and informative technologies currently available. In Specific Aim 1 we propose the first platform, a whole genome analysis of copy number changes using Illumina BeadArray(tm) technology. The Illumina 550,000 SNP BeadChip will be used to create high- resolution copy number maps of whole cancer genomes. Our side-by- side comparison of the current 300,000 SNP BeadChip to Digital Karyotyping shows that it provides higher resolution at a dramatically lower cost. Our data show that this technology can be used to accurately detect deletions and amplifications, powerful indicators of genes of interest. Specific Aim 2 will characterize the CpG island hypermethylome in human cancer. Leading experts in the field of cancer epigenetics have a proven approach to define the set of possible genes that have expression silencing as a result of hypermethylation. This assay will again use the Illumina technology for high-throughput and accurate assay of functionally selected 5'CpG islands across the genome, as well as all CpG islands located on chromosomes 21 and 22, and a random selection of non- CpG island CpG sites located on these two chromosomes. In addition to DNA mutations and structural changes, DNA methylation is a significant cause of abnormal gene silencing linked to the development of cancer. In the informatics and data verification proposed in Specific Aim 3 we have developed an integrated approach to organized transmission of raw and verified data to the NCICB Data Coordinating Center using CaBIG-compliant data feeds. We will also make available data analysis tools developed as part of this project. These include means to integrate diverse genomic data from normal and cancer genomes, user- friendly visualization tools, web portals for data sharing, and use machine learning to derive systems biology correlates among available TCGA data, all designed for CaBIG silver or better compliance. Upon successful completion of these aims, we will have a rapid and efficient means to assay tens of thousands of cancer genomes, and rapidly produce biologically meaningful data on copy number changes and hypermethylation. This project has direct relevance to public health. Precise knowledge of the type and frequency of the major cancer causing alterations will allow the best molecular targets to be selected for new therapy development.